Much has been written on the use of observational studies in epidemiological research. It will be necessary to apply many of the same epidemiologic techniques used in conventional observational studies to the clinical data bank. Work on determining which epidemiologic approaches are most appropriate for use with clinical data banks has begun in conjunction with the Stroke and Traumatic Coma Databank Networks (N01-NS-8-2309, 6, 7, 8, N01-NS-9-2302, 96, 97, 98). A presentation at the Society for Clinical Trials was based, in part, on the work of this project.